options(width=100)
knitr::opts_chunk$set(out.width='1000px',dpi=200,message=FALSE,warning=FALSE)#load packages and csv file
library(ggplot2)
library(dplyr)
library(gridExtra)
library(ggthemes)
library(RColorBrewer)
library(datamodelr)#main
active_periods<-read.csv('active_periods.csv',sep=',',stringsAsFactors=F)
battle_actors<-read.csv('battle_actors.csv',sep=',',stringsAsFactors=F)
battle_durations<-read.csv('battle_durations.csv',sep=',',stringsAsFactors=F)
battles<-read.csv('battles.csv',sep=',',stringsAsFactors=F)
battle_dyads<-read.csv('battle_dyads.csv',sep=',',stringsAsFactors=F)
commanders<-read.csv('commanders.csv',sep=',',stringsAsFactors=F)
enum_surpa<-read.csv('enum_surpa.csv',sep=',',stringsAsFactors=F)
enum_wina<-read.csv('enum_wina.csv',sep=',',stringsAsFactors=F)
#terrain
terrain<-read.csv('terrain.csv',sep=',',stringsAsFactors=F)
enum_terra1<-read.csv('enum_terra1.csv',sep=',',stringsAsFactors=F)
enum_terra2<-read.csv('enum_terra2.csv',sep=',',stringsAsFactors=F)
enum_terra3<-read.csv('enum_terra3.csv',sep=',',stringsAsFactors=F)
#weather
weather<-read.csv('weather.csv',sep=',',stringsAsFactors=F)
enum_wx1<-read.csv('enum_wx1.csv',sep=',',stringsAsFactors=F)
enum_wx2<-read.csv('enum_wx2.csv',sep=',',stringsAsFactors=F)
enum_wx3<-read.csv('enum_wx3.csv',sep=',',stringsAsFactors=F)
enum_wx4<-read.csv('enum_wx4.csv',sep=',',stringsAsFactors=F)
enum_wx5<-read.csv('enum_wx5.csv',sep=',',stringsAsFactors=F)
#belligerent
belligerents<-read.csv('belligerents.csv',sep=',',stringsAsFactors=F)
enum_reso<-read.csv('enum_reso.csv',sep=',',stringsAsFactors=F)
enum_pri<-read.csv('enum_pri.csv',sep=',',stringsAsFactors=F)Some tables are related together via primary –> foreign keys like in a SQL Db. This package(datamodelr) helps to represent these relationships in a ERD.
df_f <- dm_from_data_frames(battles, battle_actors,battle_dyads,battle_durations,active_periods,commanders,
enum_wina,enum_surpa,
belligerents,enum_pri,enum_reso,
terrain, enum_terra1, enum_terra2, enum_terra3,
weather,enum_wx1,enum_wx2,enum_wx3,enum_wx4,enum_wx5)
display <- list(accent1 = c('active_periods','battle_actors','battle_durations','battle_dyads','weather','belligerents','terrain'),
accent2 = c('enum_wina','enum_surpa'),
accent3 = c('enum_terra1','enum_terra2','enum_terra3'),
accent4 = c('enum_wx1','enum_wx2','enum_wx3','enum_wx4','enum_wx5'),
accent5 = c('enum_reso','enum_pri'),
accent6 = c('commanders'))
df_f <- dm_set_display(df_f, display)
df_f <- dm_add_references(df_f,
battle_actors$isqno == battles$isqno,
battle_durations$isqno == battles$isqno,
battle_dyads$isqno == battles$isqno,
active_periods$isqno == battles$isqno,
belligerents$isqno == battles$isqno,
weather$isqno == battles$isqno,
terrain$isqno == battles$isqno,
commanders$actors == battle_actors$actor,
battles$wina == enum_wina$value,
battles$surpa == enum_surpa$value,
belligerents$pri1 == enum_pri$value,
belligerents$pri2 == enum_pri$value,
belligerents$pri3 == enum_pri$value,
belligerents$reso1 == enum_reso$value,
belligerents$reso2 == enum_reso$value,
belligerents$reso3 == enum_reso$value,
terrain$terra1 == enum_terra1$value,
terrain$terra2 == enum_terra2$value,
terrain$terra3 == enum_terra3$value,
weather$wx1 == enum_wx1$value,
weather$wx2 == enum_wx2$value,
weather$wx3 == enum_wx3$value,
weather$wx4 == enum_wx4$value,
weather$wx5 == enum_wx5$value)
graph_df <- dm_create_graph(df_f, rankdir = "RL", col_attr = c("column", "type"),view_type = "keys_only")
dm_render_graph(graph_df)enum typebelligerents table has 6 columns (sec1,sec2,sec3,reso1,reso2,reso3) encoded with 2 enum tables
belligerents %>% select(sec1,sec2,sec3,reso1,reso2,reso3) %>% head(5)## sec1 sec2 sec3 reso1 reso2 reso3
## 1 0
## 2 RR WD
## 3 0 PS WD
## 4 FE LF RR WL
## 5 0 WDAnd the 2 enum tables look like :
enum_pri## value description
## 1 LF Left flank
## 2 DO Defensive/offensive plan
## 3 PP Penetration
## 4 RR Right rear
## 5 EE Single envelopment
## 6 DD Defensive plan
## 7 DE Double envelopment
## 8 RF Right flank
## 9 LR Left rear
## 10 FE Feint, demonstration, or holding attack
## 11 FF Frontal attack
## 12 RC River crossingenum_reso## value description
## 1 AA Annihilated
## 2 PS Pursued
## 3 PP Penetration
## 4 WD Withdrew
## 5 RR Repulse
## 6 BB Breakthrough
## 7 SS Stalemate
## 8 WL Withdrew with heavy lossesTo match both belligerents table and the enum tables :
belligerents$sec1<-sapply(belligerents$sec1, function(x) enum_pri$description[match(x, enum_pri$value)])
belligerents$sec2<-sapply(belligerents$sec2, function(x) enum_pri$description[match(x, enum_pri$value)])
belligerents$sec3<-sapply(belligerents$sec3, function(x) enum_pri$description[match(x, enum_pri$value)])
belligerents$reso1<-sapply(belligerents$reso1, function(x) enum_reso$description[match(x, enum_reso$value)])
belligerents$reso2<-sapply(belligerents$reso2, function(x) enum_reso$description[match(x, enum_reso$value)])
belligerents$reso3<-sapply(belligerents$reso3, function(x) enum_reso$description[match(x, enum_reso$value)])belligerents %>% select(sec1,sec2,sec3,reso1,reso2,reso3) %>% head(5)## sec1 sec2 sec3 reso1 reso2 reso3
## 1 <NA> <NA> <NA> <NA> <NA> <NA>
## 2 <NA> <NA> <NA> Repulse Withdrew <NA>
## 3 <NA> <NA> <NA> Pursued Withdrew <NA>
## 4 Feint, demonstration, or holding attack Left flank <NA> Repulse Withdrew with heavy losses <NA>
## 5 <NA> <NA> <NA> Withdrew <NA> <NA>weather$wx1<-sapply(weather$wx1, function(x) enum_wx1$description[match(x, enum_wx1$value)])
weather$wx2<-sapply(weather$wx2, function(x) enum_wx2$description[match(x, enum_wx2$value)])
weather$wx3<-sapply(weather$wx3, function(x) enum_wx3$description[match(x, enum_wx3$value)])
weather$wx4<-sapply(weather$wx4, function(x) enum_wx4$description[match(x, enum_wx4$value)])
weather$wx5<-sapply(weather$wx5, function(x) enum_wx5$description[match(x, enum_wx5$value)])
terrain$terra1<-sapply(terrain$terra1, function(x) enum_terra1$description[match(x, enum_terra1$value)])
terrain$terra2<-sapply(terrain$terra2, function(x) enum_terra2$description[match(x, enum_terra2$value)])
terrain$terra3<-sapply(terrain$terra3, function(x) enum_terra3$description[match(x, enum_terra3$value)]) History :